Am Mittwoch, 17. M=E4rz 2004 02:04 schrieben Sie:
> Howdy everyone, a question on the best solution to this particular
> challenge.
>
> I'm a member of a car club, and we often have track days where we're timed
> around a racetrack. Currently the timing consists of people with
> stopwatches clicking the button when they see a car go past.
>
> Needless to say this is pretty inaccurate and not very reliable.
>
> Recently someone asked me if I could make a timing system, using a
> transmitter on each car (as cheap as possible) and a "beacon" that would
> detect the passing of the car, ans also identify the car.
>
> I've thought of 2 possible solutions and would appreciate feedback on what
> which is the most suitable, and reliable, and whether anyone has any bett=
er
> ideas...
>
> The first solution would consist of an IR led on each car, flashing at a
> discrete frequency, or even flashing a predetermined "ID" code. A beacon
> would sit by the side of the track at the timing marker and count pulses,
> identifying each car in turn. In this case the beacon sees the car pass. I
> thought initially of using a discrete micro on each car, but it would be
> overkill. A 555 based timer would be easier and cheaper...
>
> This has the advantage of being reasonably cheap to implement, and it's
> easy to have a different frequency or code for all of the different cars
> (about 50 different cars per event, but only 8 on the track at any one
> time)
>
> Another option is to have an IR transmitter on the beacon, and an IR
> receiver on the car. The car would also have a RF transmitter, so when the
> car sees the beacon pass, it identifies itself via RF. This would be more
> expensive, but I think more reliable.
>
> The range from the beacon to the car would be 5 - 20 metres (15-60 feet),
> depending on what line they take through the corner. Speeds at this point
> of the track are generally 70 - 120km/hr (45-75 MPH)
>
> The angle between the beacon's line of sight and the path of the car would
> ideally be 90 degrees, but knowing some of the drivers this could vary
> considerably!
>
> The software and timing will be handled by a PC attached to the beacon (or
> communicating to the beacon through RF again) and displaying the times.
> I've done that part, just need to handle the detection and identification
> of the cars now...
>
> Any ideas anyone?
>
> --
> http://www.piclist.com hint: To leave the PICList
> mailto:piclist-unsubscribe-request@mitvma.mit.edu



I can see a problem... what if two cars do pass the finish line nearly at t=
he=20
same time just with a difference of less than a car length? I tell you, no =
IR=20
method from either the right or the lift side, or even both will work.
You need to detect a passing car from over or under the track!

So the easiest and cheapest solution that comes into my mind now is to plac=
e a=20
frame on the finish line. this frame holds a set of IR diodes in place over=
=20
the track.=20
The cars will be equipped with IR recivers turned upwards and RF transmitte=
rs=20
to send the ID signal. (the IR recivers must be placed at the front of all=
=20
cars, so they're all at the same place relative to the car, to be fair.)
When a car passes the frame and catches the IR signal it will send its ID=20
signal via RF. This signal can be catched up by a computer that does all th=
e=20
needed calculations and processing. The RF ID circuit can be a cheap digita=
l=20
circuit connected to a cheap RF transmitter module.

A more advanced and professional method would be to put RFID's under the ca=
rs=20
and the reciver units under the track. The powering reciver units hafe to b=
e=20
powered very low, to limit their range to the finish line. Everytime a car=
=20
passes the line its RFID get's powered by the reciver and sends its ID sign=
al=20
back. This ID signal gets a time stamp when recived and then you have all=20
needed data for processing...

MfG,
Do.Pe.

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu